Marine Environmental Research 22 (1987~ 167-176

Elevated Circulating Erythrocyte Micronuclei in Fishes from Contaminated Sites off Southern California Jo Ellen Hose, ~ Jeffrey N. Cross, b Steven G. Smith" & Dario Diehl b VANTUNA Research Group, Department of Biology. Occidental College, 1600 Campus Road. Los Angeles, California 90041. USA hSouthern California Coastal Water Research Project, 646 West Pacific Coast Highway, Long Beach. California 90806, USA (Received 26 October 1986; revised version received 19 December 1986; accepted 2l December 1986)

A BSTRA CT Frequencies of circulating eo'throcyte micronuclei in two rnarine fish species fi'om contaminated areas off" southern California were elevated relative to [ishes from less contaminated sites. Micronuclei frequencies from contamhtated sites were ]'ot~r times higher in white croaker (Genyonemus lineatus) and eleven times higher in kelp bass ( Paralabrax clathratus). The increased micronuclei frequency was related to previously determined environmental concentrations of chlorinated hydrocarbons (DDTs and PCBs) and poly¢3"clic aromatic hydrocarbon metabolites. However, micronuclei frequency was only weak O"correlated to individual body b,trdens of chlorinated hydrocarbons in white croaker as determined in this study. Applications and limitations of piscine micronucleus measurements are discussed.

INTRODUCTION One primary goal of aquatic toxicology is the development of animal screening techniques that measure environmental contamination. Among the most commonly used techniques are chemical contaminant, biochemical, cytogenetic and histopathological analyses (Dixon, 1985). One of the most promising, inexpensive and rapid screening techniques suitable for 167 Marine Environ. Res. 0141-1136/87/$03-50 © Elsevier Applied Science Publishers Ltd. England, 1987. Printed in Great Britain.

168

Jo Ellen Hose, Jeffrey ,V Cross, Steven G. Smith, Dario Diehl

evaluating exposure to contaminants for marine and freshwater fishes is the micronucleus test (Landoh & Kocan. 1983). Micronuclei are smaller, secondary nuclei formed following chromosomal breakage (Schmid, 1976). Although micronuclei may arise spontaneously, induction of micronuclei is commonly used to detect genotoxic damage resulting from mutagen exposure (Heddle et al.. 1983). Originally developed as a test using rodent polychromatic erythrocytes in bone marrow and later extended to circulating erythrocytes (MacGregor et c~l., 1980), the micronucleus test has also been applied to nucleated piscine erythrocytes (Hooftman & de Raat, 1982; Hose et al., 1984). Although all the causes of micronucleus formation in fish are not known, micronuclei have been induced following continuous exposure to ethyl methanesulfonate (Hooftman & de Raat, 1982) and benzo(a)pyrene (Hose et al., 1984). In both cases the incidence of micronucleus formation was dosedependent, Hooftman & de Raat (1982) observed the presence of additional anomalies such as the loss of the usual elliptical shape ofthe nucleus and the appearance of irregular Feulgen-positive structures in the cytoplasm. Results from these two studies indicate that the piscine micronucleus test may be a simple and rapid alternative to the more routinely used cytogenetic tests such as the occurrence of chromosomal aberrations and sister chromatid exchanges. Since most marine fishes have numerous small chromosomes, cytogenetic methods that require visualization of individual chromosomes are difficult and time-consuming (Landolt & Kocan, 1983). The micronucleus test is applied independent of karyotypic characteristics and is equally suitable for any fish species. The objective of this study was to determine if the incidence of micronuclei in circulating erythrocytes was elevated in fishes from a highly contaminated environment. White croaker (Sciaenidae: G e n y o n e m u s lineatus) and kelp bass (Serranidae: P a r a l a b r a x clathratus) were chosen for this study because they are important local sportfishes (Fitch & Lavenberg, 1971: Puffer et al., 1982) and they respectively occupy increasing trophic levels in the marine food web (Mearns & Young, 1980). White croaker reside near the sediment-water interface and consume a variety of benthic and epibenthic organisms. Because white croaker from portions of San Pedro and Santa Monica bays have high concentrations of DDTs and PCBs (Brown et al., 1982, 1986), human consumption warnings have been posted in several areas. Kelp bass live in kelp beds and rocky reefs and feed on small fish, including young croaker, and invertebrates; frequently they contain high concentrations of chlorinated hydrocarbons (Gossett et al., 1982). In this study, frequencies of erythrocyte micronuclei were compared between fishes collected from the highly contaminated San Pedro Bay and from less contaminated sites away from metropolitan Los Angeles. The San

169

Micronuc[ei in southern California fishes

Pedro Bay-White Point area received enormous amounts of D D T from municipal wastewater discharge until the early 1970s iSchafer, 1984). Current levels of D D T and PCB in sediment exceed those measured in Commencement Bay, Washington (Brown et al.. 1986), a federal Superfund site (NOAA, 1986). San Pedro Bay also receives significant inputs of polycyclic aromatic hydrocarbons from natural oil seeps, shipping activity and the Los Angeles River (Gossett el al., 1983).

METHODS

metropolitan Los Angeles and from reference sites 40 to 80 km away (Fig. l I. White croaker were obtained from outer Los Angeles Harbor in San Pedro Bay and off Dana Point. Kelp bass were collected from White Point on the Palos Verdes Peninsula and off the west end of Santa Catalina Island. Fishes were taken to the laboratory and acclimatized for two to three weeks prior to sampling. They were anaesthetized with methyl tricaine sulfonate. Using cardiac puncture, approximately I cc o f blood was 118~.(}" i

POINT DUME

~':.,

11a o

.

],

I~

2.0

LOSA~G.. ~4Q

LOS ANGELES CITY OUTFALL " ~ Sonto Moni¢a Bay

: /'~-.-

REOONO0 BEACH

LOSANGELESCOUNTY OUTFALL

~-'

13ALOS VER{)ES PENINSULA

Pedro Boy

~.~....:..

" ~ ' ~ - ~'.-~i:" •

ORANGE COUNTY OUTF,:,LL

SANTA CATALINA ISLAND

'

LK R * ~ 33°20 '

Fig. 1. Map of fish collection sites in southern California. WC = contaminated site (outer Los Angeles Harbor) for white croaker; WR = reference site (Dana Pointl for white croaker; KC = contaminated site (White Point) for kelp bass: KR = referencesite (Catalina Island) for kelp bass. Scale bar is calibrated in miles.

170

Jo Ellen Hose. Jell?e)" N. Cross, Steven G. Smith, Dario Diehl

withdrawn into a heparinized 1-cc tuberculin syringe fitted with a 21-G needle. Blood smears were immediately prepared and fixed in absolute methanol for 15min. Smears were then stained with May-Grunwald Giemsa. Blood smears were examined microscopically under high ( 1000 x ) power. Numbers of micronucleated erythrocytes (MN) per 1000 erythrocytes were determined on coded slides and scored using blind review by a single observer. The number of micronucleated erythrocytes was the average of two determinations and was expressed as MN%o. Selected fishes were analyzed for total D D T (DDE + D D D + DDT) and total PCB (Aroclor 1 2 4 2 + A r o c l o r 1254). Methods for chlorinated hydrocarbon contaminant analysis follow Gossett et al. (1983). Liver samples (1 to 5 g) were homogenized in a 20-ml aliquot of pesticide-quality acetone, filtered and re-extracted into n-hexane. The hexane fraction was prepared by cleaning on activated Florisil (750~'C for 4h). Analysis for chlorinated hydrocarbons was performed using a Tracor MT220 gas chromatograph (GC) equipped with an electron-capture detector and a l-8m x 2 m m i . d . glass column packed with 1-5% OV17+ 1"95% QFI on 80 to 100 mesh Gaschrom Q. Column temperature was 200°C with a nitrogen flow of 20 ml/min.

RESULTS Frequencies of MN were elevated fourfold in white croaker from San Pedro Bay and elevenfold in kelp bass from White Point compared to less contaminated sites (Table 1). The raw data were transformed to log to (x + 1) (where x = MN%0 because the data fit a negative binomial distribution (variance greater than the mean) and the variances of the contaminated fishes were significantly greater than those of the reference fishes. T-tests of the log-transformed data showed that the incidences of MN were significantly higher in white croaker and kelp bass from contaminated sites than those of the reference sites (white croaker: t = 6'384, d f = 54,p < 0.001; kelp bass: t = 7.511, d r = 28, p < 0-001). Micronuclei counts were significantly higher in kelp bass than in white croaker at the contaminated sites (t = 2.743, d f = 41,p < 0-01) but not at the reference sites (t = 0.155, d f = 41, p > 0.50). There was no difference between MN frequencies in male and female white croaker at either site. Piscine erythrocytes are elliptical with central elliptical nuclei (Fig. 2). Micronuclei observed in this study usually ranged in diameter from 1/20 to 1/10 of that of the erythrocyte nucleus; rarely M N approaching 1/3 the size of the parent nucleus were found. Staining characteristics of nuclei and MN

MicronucIei in southern Catl~brnia fishes

~71

TABLE 1 Frequency of Micronucleated Erythrocytes in Peripheral Circulation of Fishes from Contaminated and Reference Sites in Southern California (values are the mean number of micronucleated erythrocytes per t000 ceils [M~No,~1; SD = standard deviation: n = sample size) Species

Contaminated site ~ (..~[V%~ + SD (n))

R~}ference site ~ (MN%~ + S D (n))

White croaker tog~ o-transformed

34 + 2 7 (28l 0.59 + 0.24{ 28)

0-8 + l - t (28i 0' 19 + 0.22 (28}

Kelp bass Ioglo-transformed

6'8 +5.1 {15} 0-81 +0-27 (15)

0"6 +0"6 {15i 0-18 + 0.16(15j

Contaminated site t'or white croaker = San Pedro Bay; contaminated site for kelp bass = Palos Verdes Peninsula. "Reference site for white croaker = Dana Point; reference site for kelp bass = Catalina Island.

W Fig, 2, Erythrocytes from white croaker (Genyonemus lineatus) caught at reference site off Dana Point, California. Nuclei are elliptical and rarely contain micronuclei. May-Grunwald Giemsa. 3000 ×.

t72

b

Jo Ellen Hose, Jt:8-rey ),, Cross, Steren G. Smith. D,,lrio Dieh-

i

Fig. 3. Micronucleated erythrocytes {arrows) of white croaker from contaminated San Pedro Bay site. Micronuctei are visible as attached, knoblike structures [single a.~ow) or large detached secondary nuclei !double arrowl. Note extreme nuclear pteomorphism in this fish. Small dark dots in cytoplasm are bacteria. They are easily distinguished from isolated micronuctei because of their refractile nature and their location on the cell surface. Bacteria were identified using a Gram stain. May-Grunwald Giemsa. 3000 ×

are similar. Two types of MN were observed: attached, knobtike nuclear segments (Fig. 31)resembling those described by Hooftman & de Raat (1982) and isolated nuclear fragments identical to MN in mammalian cells. Although the attached nuclear fragments do not conform to the classical definition of a micronucleus in a mammalian celt (Schmid, 1976), they are quantifiable manit'estations of genotoxicity in lower vertebrates (Hoot'tman & de Raat, 1982) and hence have been included in the MN counts conducted in this study. M N frequencies in reference fishes, averaging from 0-6 to 0"8%0, are lower than the spontaneous M N frequency of l to 3%0 reported for mice polychromatic erythrocytes (Heddle et al.. 1983). Nuclear pleomorphism, also mentioned by Hooftman & de Raat (1982), was found in this study in smears with high MN frequencies. Chlorinated hydrocarbon contaminants were at least ten times higher in fishes from contaminated sites when compared to those from the reference sites (Table 2). Correlations between log-transformed MN frequencies and individual log-transformed total D D T (r = 0-156, d f = 34, 0-20 < p < 0'50) and total PCB (r = 0-186, d f = 34, 0"20 < p < 0"50) in white croaker were not significant.

MicronucIei in southern Cal!l~)rnia fishes

173

TABLE 2 Hepatic Contaminant Concentrations in Fishes from Contaminated and Reference Sites {values are in ppm ~et weight: ,g= mean: SD = standard deviation: n = sample sizel

Species

Contaminated site" Total DDT (A'+SD(n))

White croaker Kelpbass

Total PCB (X+SD(n))

3.13 + 5"441281 1-36+0.98(28) 8.28+7.37 (5) 2.57+3-I8 (5)

Reference site ~ Total DDT ()(+SD(n))

Tom/PCB (.Y-SD(n))

0.17 +0.07(8) 0-85+0'31(51

0-03---0-05(8) 0.42-,-0"I8(5t

"Contaminated site for white croaker = San Pedro Bay; contaminated site for kelp bass = Palos Verdes Peninsula. Reference site tbr ~vhite croaker = Dana Point; reference site for kelp bass = Catalina Island.

DISCUSSION Genotoxic damage was greater in fishes from highly contaminated San Pedro Bay compared to fishes collected at Dana Point and Catalina Island. Kelp bass, which generally had higher body burdens of chlorinated hydrocarbons than white croaker, also had higher frequencies of MN. Although factors which modify MN induction are poorly understood and may include factors such as blood cell kinetics, temperature, life history stage and possibly sex differences (Heddle et al., 1983), results obtained in this study do support the overall relationship between contaminant exposure and in vivo genotoxicity. Induced spawning experiments have shown that maternal MN frequencies in white croaker were predictive of reproductive success. Eggs from fish with elevated MN counts had lower fertilization rates (Cross & Hose, 1986). Micronucleus counts in white croaker were .only weakly correlated with body burdens of total DDTs and PCBs; these results are consistent with the non-clastogenic properties of these substances (Heddle et al., 1983). San Pedro Bay white croaker body burdens of oxygenated metabolites of DDT and PCB (primarily hydroxylated and conjugated derivatives) are up to 60 times the concentrations of the parent compounds (Brown et al., 1982). The mutagenic potential of the chlorinated hydrocarbon metabolites has not been adequately defined. Other potential clastogens, including chlorinated benzenes, have been found in San Pedro Bay fishes (Young et al., 1980; Malins et al., 1986). Although polycyclic aromatic hydrocarbons (many of which are well-known mutagens) were not measured in this study, previous work has shown that San Pedro Bay sediments contain high (up to 18 ppm dry weight) concentrations of benzo(a)pyrene (Gossett e¢ al., 1983). Biliary

174

Jo Ellen Hose, Jeffrey .~: Cross, Steven G. Smith, Dario Diehl

polycyclic aromatic hydrocarbon metabolites were highest in white croakers from outer Los Angeles Harbor (3"70 _+ 3-10 p p m wet weight, X_+ SD, n = 7), less off White Point (0-96 +_ l'60ppm, n = 12) and lowest off Dana Point (0"072 + 0-074ppm, n = 9) (Malins et al., 1986). The piscine MN test may be a more sensitive indicator of chronic, rather than acute, exposure to contaminants. After a three-week exposure. Hooftman & de Raat (1982) did not observe M N induction in Umbra pygmaea except at 200mg/liter ethyl methanesulphonate. Micronucleus formation was evident at their lowest dosage of 8 rag/liter after six weeks. The time-dependent response could limit the use of the piscine MN test to monitor chronic pollution rather than as an early warning system. Mammalian studies suggest that the in vivo M N test is less sensitive than the Ames test due to the lack of activated metabolites reaching target bone marrow cells (Natarajan & Obe, 1986). The piscine MN test does not have this limitation since head kidney (the major site of hematopoiesis in adult fish) contains high mixed function oxygenase activity (James et al., 1979a,b). Formation of MN can also be measured in larval fishes by preparing squash-smear preparations of liver (Hose et al., 1984), which is the hematopoietic organ during larval stages. This technique can be applied to archived, formalin-fixed specimens. Results of this study support the use of the piscine micronucleus test as a rapid monitoring tool to detect the presence of genotoxic agents in the environment. A similar system using circulating erythrocytes has been proposed for use in humans (MacGregor et al., 1980). Recently, two promising M N models, the newt (Pleurodeles waltl) (Grinfield et al., 1986) and the tadpole (Rana catesbeiana) (Krauter et al., in press), have been described for assessing contaminants in freshwater. The simplicity, rapidity and biological relevance of the piscine MN test suggest that further laboratory validation studies are warranted. Background information on fish erythrocyte kinetics, persistence of micronucleated erythrocytes, and interspecific and environmental variables that affect MN formation are necessary for the development of an in vivo monitoring system based on MN frequencies.

ACKNOWLEDGEMENTS We thank H. Parker and D. DiFiore for their help..H. Stubbs and R Hershelman assisted the field collections. Laboratory facilities were provided by Southern California Edison. The study was supported by National Oceanic and Atmospheric Administration Grant No. NA-85ABD-00003.

M icronuclei in southern Cal(forniafishes

175

REFERENCES Brown, D. A., Gossett, R, W. & Jenkins, K. D. (1982). Contaminants in white croakers Genyonemus lineatus (Ayres, 1855) from the Southern California Bight: II. Chlorinated hydrocarbon detoxification, toxification. In: Physiological mechanisms of marine pollutant toxicity (Vernberg, W, B., Calabrese, A., Thurberg, F. P. & Vernberg, F. J. (Eds)), New York, Academic Press, 197-213. Brown, D. A., Gossett, R. W., Hershelman, G. P., Ward, C. F., Wescott, A. M, & Cross, J. N. (1986). Municipal wastewater contamination in the Southern California Bight: Part I--Metal and organic contaminants in sediments and organisms. Mar. Environ. Res., 18, 291-310. Cross, J. N. & Hose, J. E. (19861. The reproductive cycle ofdemersal fishes in an area receiving urban wastes (Abstract). In: Sixth International Ocean Disposal Symposium, April 21-25, 1986, Pacific Grove. Calijbrnia, 116-17. Dixon, D. R. (1985). Cytogenetic procedures. In: The effects of stress and pollution on marine animals (Bayne, B. L., Brown, D. A., Burns, K., Dixon, D. R., Ivanovici, A., Livingstone, D. R., Lowe, D. M., Moore, M. N.. Stebbing, A. R. D. & Widdows, J. (Eds)), New York, Praeger Scientific, 205-14. Fitch, J. E. & Lavenberg, R. J. (I 97 I). Marine food and game fishes of California, Berkeley, University of California Press. Gossett, R. W., Puffer, H. W., Arthur, R. H., Alfafara, J. H. & Young, D. R. (1982). Levels of trace organic compounds in sportfish from Southern California. In: Coastal Water Research Project biennial report 1981-1982 (Bascom, W. (Ed.)), Long Beach, California, SCCWRP, 29-37. Gossett, R. W., Puffer, H. W., Arthur, R. H., Jr & Young, D. R, (1983). DDT, PCB and benzo(a)pyrene levels in white croaker (Gen.vonemus lineatus) from Southern California. Mar. Pollut. Bull., 14(2), 60--5. Grinfield, S., Jaylet, A., Siboulet, R., Deparis. R & Chouroulinkov, I. (1986). M icronuclei in red blood cells of the newt Pleurodeles waltl after treatment with benzo(a)pyrene: Dependence on dose, length of exposure, post-treatment time, and uptake of the drug. Environ. Mutagen., 8(t), 41-51. Heddle, J. A., Hire, M., Kirkhart, B., Mavourin, K., MacGregor, J. 7"., Newell, G. W. & Salamone, M. F. (1983). The induction of micronuclei as a measure of genotoxicity. Mutation Res., 123, 61-118. Hooftman, R. N. & de Raat, W. K. (1982). Induction of nuclear anomalies (micronuclei) in the peripheral blood erythrocytes of the eastern mudminnow (Umbra pygmaea) by ethyl methanesulphonate. Mutation Res., 104. 147-52. Hose, J. E., Hannah, J. B., Puffer, H.W. & Landolt, M. L. (1984). Histologic and skeletal abnormalities in benzo(a)pyrene-treated rainbow trout alevins. Arch. Environ. Contain. ToxicoL, 13(1), 675-84. James, M. O., Bowen, E. R., Dansette, P. M. & Bend, J. R. (1979a). Epoxide hydrase and glutathione S-transferase activities with selected alkene and arene oxides in several marine species. Chem.-Biol. b~teract., 25, 32144. James, M. O., Khan, M. A. Q. & Bend, J. R. (1979b). Hepatic microsomal mixedfunction oxidase activities in several marine species common to coastal Florida. Comp. Biochem. Physiol., 62C, 155-64. Krauter, P. W., Anderson, S. L. & Harrison, F. L. (in press), Radiation-induced micronuclei in peripheral blood erythrocytes of Rana catesbeiana: An aquatic animal model for in vivo genotoxicity studies. Environ. Mutagen.

176

Jo Ellen Hose. Jeffrey ),: Cross, Steven G. Srntth. Dario Diehl

Landolt. M. L. & Kocan, R. M. (1983). Fish cell cytogenetics: A measure of the genotoxic effects of environmental pollutants. In: Aquatic toxicology tNriagu, J. R. (Ed,)), New York, John Wiley and Sons, 336-53. MacGregor. J. T., Wehr, C. M. & Gould, D. H. i1980i. Clastogen-induced micronuclei in peripheral blood erythrocytes: The basis of an improved micronucleus test. Environ. Mutagen., 2(4), 509-14. Malins, D C., McCain, B. B., Brown, D. W., Myers, M. S. & Chan, S.-L. i1986). Marine pollution study: Los Angeles viciniO'. Final report to California State Water Resources Control Board. Seattle, Washington, NOAA, US Dept of Commerce. Mearns, A. J. & Young. D. R. (1980). Trophic structure and pollutant flow in a harbor ecosystem. In: Coastal Water Research Project biennial report 1979-1980 (Bascom, W. (Ed.)), Long Beach, California, SCCWRP. 287-98. Natarajan, A. T. & Obe, G. (1986). How do in t'ivo mammalian assays compare to in vitro assays in their ability to detect mutagens? Mutation Res., 167, 189-201. NOAA (1986). The national status and trends program for marine environmental quality. Inventory of chlorinated pesticide and PCB data for US marine and estuarine fish and invertebrates. Ocean Assessments Division. Seattle, Washington. NOAA, US Dept of Commerce. Puffer, H. W., Azen, S. P. & Duda, M. J. (1982). Sportfishing activity and catches in polluted coastal regions of metropolitan Los Angeles. ~\: .4m. J. Fish. Mgmt. 2, 75-9. Schafer, H. A. (1984). Characteristics of municipal wastewater. In: Coastal Water Research Project biennial report 1983-1984 (Bascom, W. (Ed.)), Long Beach. California, SCCWRP, 11-19. Schmid, W. (1976). The micronucleus test for cytogenetic analysis. In: Chemical mutagens: Principles and methods for their detection, Vol. 6 (Hollaender, A. (Ed.)), New York, Plenum Press, 31-53. Young, D. R., Jan, T.-K. & Gossett, R. (1980). Chlorinated benzenes in sediment and organisms. In: Coastal Water Research Project biennial report 1979-1980 (Bascom, W. (Ed.)), Long Beach, California, SCCWRP, 181-95.